THE CON C E P T 0 FED U CAT ION
Dem1is Hayes r'1A (Kent)
This thesis is presented in fulfillr.1ent of the requirenent of the degree of Docter of Philosophy of the University of Lo~Qon.
University of London Institut~:: of Lducntion
A B S T
R
ACT
A misapprehension of the concept of education is documented.
This reflects a weakness in our grasp of the concept of the concept
of education and a faulty understanding of the nature of thought and
argument. This misapprehension involves the essentially Kantian
view of examples as intuitive aids to understanding. Against this
it
16 argued that examploo a.re nbaolutl:lly fundClffiental to thought.
Even deduction, our paradigm of good rea.soning, reduces to
case by
case a.rgument. This misapprehension is embodied in 'criterioloeical'
approaches to the concept which assume that to be justified in
calling something an example of 'education' necessitates giving
'criteria'. But justification may take the form of case by case
'argument'. This alone can give increased grasp of a concept.
The 'criteriological' approach derives from Wittgenstein. The
notion of 'rules' of language functioning as intermediaries between
'criteria' and 'conditions' for the use of terms provides no support
for the approach. There are no 'rules' in the special sense
required. A reading of Wittgenstein suggests no 'theory of criteria'
but. a methodological injunction to consider the details of what it
makes sense to say in particular cases. Three interrelated metaphors
'family resemblance', 'games' and 'grammar' have been misinterpreted
by being interpreted systematically. They are intimately connected
with case by case procedure and offer an essentialism without
universals.
A 'family resemblance' approach to the concept is developed and
discussed. The notion of 'aspect change' is employed to illuminate
some examples of education. A case by case procedure is sketched
utilising examples from Tolstoy, Dickens, Golding, Austin and others.
'Criteria' extracted from these examples would be clumsy and
CON TEN T S
Chapter One
Il'IJTRODUCTION •••••••••••••••••••••••••••••••• 1.1. 1.2.1.3.
1.4.
1.5.1.6.
1.7.
The Kantian View of Examples •••••••••••••••••••••••
Documentation
(A) •••••••••••••••••••••••••••••••••Documentation
(B) •••••••••••••••••••••••••••••••••Examples and Counter-Examples ••••••••••••••••••••••
Case by Case Argument ••••••••••••••••••••••••••••••
'Concepts' and 'Analysis' ••••••••••••••••••••••••••
Summary: 'What it makes sense to say' ••••••••••••••
Chapter Two
RULES
AND CRITERIA ••••••••••••••••••••••••••
2.1.,2.2.
2.3.
2.4.
2.5.
2.6.
2.7.
2.8.
Necessary and Sufficient Conditions ••••••••••••••••
Rules ••••••••••••••••••••••••••••••••••••••••••••••'Criteria' as Necessary and Sufficient Conditions ••
'Criteria in Philosophy of Education •••••••••••••••
The Basic Criteriological View •••••••••••••••••••••
Meaning'~ and Criteria •••••••••••••••••••••••••••••••'Criteria' in i'Jittgenstein's Philosophy ••••••••••••
General Criteria and Education •••••••••••••••••••••
Chapter Three
THREE
~ffiTAPHORS•••••••••••••••••••••••••••••
3.1.
3.2.
3.3.
1 Family Resemblance' ••••••••••••••••••••••••••. _ ••••
'Games' ••••••••••••••••••••••••••••••••••••••••••••
'Grammar' ••••••••••••••••••••••••••••••••••••••••••
Chapter Four
ESSENCE AND EDUCATION •••••••••••••••••••••••
4.1.
4.2.
4.3.
A Family Resemblance Approach? ••••••••••••••••••••
Learning and Aspect Change •••••••••••••••••••••••••
'Grammatical'
Remarks ••••••••••••••••••••••••••••••p.
p. p.5
5
7
p. 15
p. 17
p.
25
p.43
p.52
p.
54
p.55
p. 59
p. 69
p. 76
p. 81
Chapter Five
EDUCATION BY EXAMPLES •••••••••••••••••••••••
5.1.
5.2.
5.3.
The Particularity of Criticism •••••••••••••••••••••
A Case by Case Procedure •••••••••••••••••••••••••••
Some Objections ••••••••••••••••••••••••••••••••••••Chapter Six
CONCLUSION ••••••••••••••••••••••••••••••••••6.1.
6.2.
Breaking a Habit of Thought ••••••••••••••••••••••••
Wittgenstein's Full Stop •••••••••••••••••••••••••••
BIBLIOGRAPHY ••••••••••••••••••••••••••••••••••••••••••••••••
p.183
p.183
p.187
p.19Jf~p.20'
p.210
p.212.
Chapter One INTRODUCTION
, "You must'nt tell us about the ring, here. Very well, then. Describe your father as a horsebreaker. He doctors sick horses, I dare say?"
!lOh yes, Sir."
"Very well, then. He is a veterinary surgeon, a farrier, and horse breaker. Give me your definition of a horse."
(Sissy Jupe thrown into the greatest alarm by this demand.) "Girl number twenty unable to define a horse!" said Nr Gradgrind, for the general behoof of all the little pitchers.
"Girl number twenty possessed of no facts, in reference to one of the commonest of animals! Some boy's definition of a horse. Bitzer, yours." ••••••••••••
!IQuadruped. Graminivorous. Forty teeth, namely, bventy -four grinders, four eye-teeth, and twelve incisive. Sheds coat in spring; in marshy countries, sheds hoofs, too. Hoofs hard, but requiring to be shod with iron. Age known by marks in mouth." Thus (and much more) Bitzer.
"Now girl number twenty, II said fir Gradgrind. "You know what a horse is.'" (1)
Dickens Hard Times
1.1. The Kantian View of Examples.
'Certainty and clearness', according to Kant, are the two
essential requirements of critical philosophy. To meet the demand
of certainty he considered it nece3sary to reject all modes of
opining. To meet the demand of clearness it was necessary to
achieve a discursive (or logical) clearness through concepts
and an 'intuitive (aesthetic) clearness through intuition, that
is, through examples and other concrete illustrations.' (2) Intuitive clearness, for Kant, is a subsidiary business which is
only necessary 'from a popular point of view.'
(3)
Examples are merely illustrative and in giving too many vie may' cover overand conceal the articulation and organisation of the system,
which, if we are to be able to judge of its unity and solidity
( 4)
are "/hat chiefly concern us.' In these passages Kant is not
merely mru~ing eXEUses for the paucity of examples in the Critique. l/,'hat he says in the First Introduction is quite compatible with
his view of examples in enabling us to make judgements. He saw
the understanding as the 'faculty of rules' and judgment as
to see whether something is a case of something else. It is
a 'natural power' a 'mother wit' which cannot be assisted by
logic. Logic, by definition, only gives us general rules, and if
we cannot apply tllem 'logic' cannot help us. If we are wanting
in judgement Kant believes that there are two possible reasons
for this. Firstly we may have an abstract power of comprehension
which finds ap~lication to particular instances difficult. Secondly, we may not have received 'through examples and actual
practice adequate training for this particular act of judgement.'
Such a 'shar~ening o£ the judgement is indeed one great benefit
(6)
(7)
of examples.' There are, however, great drawbacks to examples.
They impair the correctness and precision of intellectual insight
because they do not enable us to comprehend the uni versali ty of rules.
'Examples are thus the go cart of judgement; and those who are
lacking in natural talent can never dispense with them.'
(8)
I begin an essay on the concept of education \vi th a briefexegesis of a very minor theme of the Critique not because I think
Kant's arguments are important or sound, for they are neither.
(9)
Nor because I believe that contemporary philosophy of education isexplicity neo-Kantian, but because in the passages I have referred
to he makes explicit the narrow conception he has about the role
of examples in thought. This attitude, as I will show, is implicit
in much of the recent work done on the philosophy of education and
vitiates all of the recent work done on the concept of education.
Crudely put, the central arguoent of this thesis is that what Kant
says is an inversion. Examples are absolutely fundamental to ( 10)
thought.
What I shall venture to call the 'Kantian view of examples'
is not absolutely wrong but it is inadequate. It is a philosophical
malaise brought about by the search for a systematic philosophy
and a philosophic system. Educational philosophers like other
philosophers have generally given up the search for philosophical
systems. However, they s'cill feel the pull of thecnalogy between
what they are doing and the system building activities of scientists.
Perhaps many educational philosophers would repudiate this analogy
in favour of one vii th literature and the arts. (11) The documentation
I "',ill present in this th8sis v/ill show any such repudiation to be
The inadequacy of the Kantian view of examples leads to an
inadequate notion of what it is to think, to reflect, to argue.
It leads to an inadequate notion of Hhat is involved in 'having'
a concept and hence to an inadequate treatment of any particular
concept. In discussing the concept of education in the light of
this malaise it is important to recognise that the inadequate
apprehension of the concept of a concept which is symptomatic of
an inadequate apprehension of the nature of thought and argument.
Of course, to say that something is inadequately apprehended is not
to say that it is not apprehended.
1.2. Documention (p.)
To document my claim that the concept of education has been
inadequately apprehended I adduce the persistence of the following
two related sets of 'questions' and the corresponding affirmative,
negative, or contrary 'answers' to them:
. (12)
(a) Is there one concept of eductlon? Is there
concept of education?
(13)
Are there two concents ofAre there three, four, or five concepts of education?
one aP.1biguous
education? (14) (15) Are there (16)
many concepts of education? Do different people have different
concepts of education? (17) Do particular social groups have different concepts of education? (18)
~1ust
every human being necessarily have a certain concept of, education? (19) Are there historically independent concepts of education? (20) Is there asuper concept of education er.1bracing possibly incompatible concepts
( 2 1 ) . (22)
of education? Is educatlon a vague concept? Is there an
essence to the concent of education? (23) Is education a family
,. ( 2 4 ) . ( 25 )
resef.1blance concept? Is educatlon a core concept? Is
.education an essentially contested concept? (26)
(b) Is it possible to define the term 'education'?
several 'definitions' of the term 'education'? (28)
(27)
Are thereAre there
necessary and sufficient conditions for the use of the term
'education'? (29) Are there necessary conciitions for the use of . (30)
the term'educatlon'? Are there sufficient but not necessary
conditions for the use of the term 'education'?
(31)
Are there anysort of conditions for the correct application of the term
'education'? (32) Are there criteria for the application of the
term 'education'?
(33)
Has the term 'education' evaluativemeaning?
(34)
Has the tero 'education' descrintive meaning?(35)
Has the term 'education' changed in meaning?
(37)
Is the meaningof the term vague? (38)
This seemingly straightforward concatenation of interrogatives
is meant to 1:e indicative of the sorts of 'questions' asked by
philosophers of education during the last decade or so. It is not
exhaustive as it ignores such 'refinements' as talk of 'strong' or
'weak' definition. That it reveals a certain amount of cOLfusion
might be more apparent if the inscriptions are rewritten in declarative
form i.e. 'There is ~ concept of education.'; 'There are two concepts of education', and so on. Hany pairings of questions from both
sections \rill seem to be incompatible if not inconsistent. Occasionally
an incompatible pairing is found in a single treatment of the concept
of educ.3tion. (39) Incidental, or even thorough going inconsistency
is of marginal importance. Hhat I lrrish to stress at the outset is
sinply the persistence of these 'questions' and of 'questions' about
these 'questions'.
Though my lists are restricted to 'questions' about the 'form' of
the concept or term a similar list could be drawn up which covered the (40)
'content' of the concept or term. Reference would be made, for
example, to philosophers who cmcnected the cO:lcept essentially with
learning and those who argue that this relation is contingent, and to
philosophers who consider 'education' to have an_~aluative component and to those who deny this. For this purpose of preparatory
documentation the list will suffice. \-Jhat they show is that something
is wrong in the 1rlay that phiJosophers talk about the concept of the
concept of education. To say that something is vlrong and to give
evidence that a certain amount of confusion exists is not, of course,
to establish that this confusion is a result of the Kantian as,sumptions
discussed above. It way be that these 'questions' are held to be
meaningful out that the correct answers to them have not been
Satisfactorily given. It will be my aim in the balance of this thesis
to establish that such continual and repeated questionings are
sterile and fruitless, because they embody the Kantian view of examples.
In correcting this Kantian misapprehension I hope that the need to ask
such 'questions' ;tiill disappear and be replaced by an approach to the
Set (a) contains examples of 'questions' about the concept of
education and the concept of the concept of education. Set (b)
contains 'questions' about the meaning of the term 'education'.
The misapprehension I am referring to is not a trivial one which
would occur if sets (a) and (b) vlere somehow confused. John
'\vilson is intolerant of phiJosophers of education who talk in an
.
(41)
'incoherent' way of both 'the use of concepts' and 'the use of terms'.
In philosophy and in ordinary conversation we move from one expression
to the other without qualm. The distinction doesn't matter unless
you are concerned with distinguishing between the nature of concepts
and the use of terms. \vilson considers such loose speaking as
important because he sees it as resulting from a tension between a
personal possessive (or 'psychological') use of the term 'concept'
and the meaning of any given term. He argues that there is no
intelligible first person posse~ve use of the term 'concept' so that all we are left with is the meaning of a term. The
misap-prehension highlighted here is wide ranging enough to cover the
confusion involved in seeing concepts as a half way house between
some sort of mental entity and the meaning of terms, and the
confusion involved in the simplistic reductionisn: of the Wilsonian
kind which involves in a search for the meaning of terms. (42) One
might also add that the misapprehension is also evinced by those
who argue that viilson has a non-dynamic view of language, a fido-fido
theory of meaning, and that his approach amounts to an attempt to
kill discussion by definition. (43)
vii Is on 's suggestion that we encapsulate bits of the world as
we distinguish it in sets of rules and that these constitute our
'concept' is reminiscent of Kant's position. For both philosophers
categorising is forming 'rules'. Philosophical argument consists of
attempts to make meanings or sets of 'rules' explicit, or it consists
of resolving categorisation problems such as asking if A is a case of
X or a case of Y, or it evaluates concepts i.e. it asks whether the
concept Y is needed. The 'grasp' of such 'rules' is not an issue.
Both philosophers would arG,lle that that is a 'psychological' r:latter.
'Laying out cases' is merely a preliminary procedure to conceptual
(44)
clarification, a guide to those lacl:ing in 'mother wit'. viilson,
therefore, exemplifies the Kantia~ view of examples. To show that his
views are not eccentric or atypical it is necessary to adumbrate the
way in which the Kantian view finds expression in the arguments of
Consideration of Hilson's argument clearly reveals that
philosophers of education are active on the three levels indicated
above. Indeed this could not be othenJise. They are concerned with
the concept of education, with the concept of that concept and with
the nature of thought and argument. To begin to understand the
nature of the Kantian misapprehension the least concrete level must
be considered first.
The following characterisation is not given a'.priori. Nor does
the presence of any or all of the following features do more than
indicate the adoption of the Kantian view. They proceed from the
most general (often merely implied) features to the slightly less
general:
1) The assumption that the only tru1y valid form of reasoning is that which approximates to deduction.
2) The belief that there is one single feature or set of features that che.racterise a concept or give the meaning of a term.
3) An excessivep-eoccupation with definition.
4)
The production of 'universal' or 'conceptual truths'.5)
An obsession with giving 'rules', 'criteria', or (logically) necessary conditions.6)
An on-going argument about method.7)
A constant need to reformulate analyses.8)
A dearth of examples.This latter feature is ex hypothesi the most obvious. 1) and 2) give expression to the pull of the analogies with mathematics and
science. All these features result from a concern 1r/i th the general
rather than the particular. They are the manifestations of certain
confusions or tendencies with a con~on root. Familiarity 1rath them make an insistence on documenting their appearance in the arguments
of philosophers of education seem otiose. But these simply stated
points are not merely the well knoiom truisms of a plethora of
introductory textbooks. They are evidence of a flawed apprehension
of what i t is to have a concept. Familiarity \.Ji th their denials
may make a persistence in documenting them appear perverse.
Assertio~denial, reformulation and endless qualification is
precisely what is being' sighted as evidence of a flawed apprehension.
Or rather, they are evidence of a flaHed apprehension of a flawed
apprehension of a flaued apprehension of the concept of the concept of
education, which is the result of a flawed apprehension of a flawed
apprehension of the nature of thought and argument. TheBe iterations
indicate the complexity of the matter. ~hese tendencies, like the 'truism' that man is mortal and must therefore die are both known
and not known.
My first point is the easiest and the hardest to document. It is
the easiest because the presence of all the other features establish
it. It is the hardest because the contradictory thing is often
asserted. Occasionally philosophers will reveal themselves in
off-hand comments. Mary ',Jarnock, for example, in her article 'Towards a
Defini tion of '~;uali ty mn Education' begins ".,ri th a remark to the effect
that 'It must be agreed from the start that in this particular field
.
(46)
there is no such thJ.ng as proof.' Ivarnock gives us a picture of
proof ldhich we could encapsulate in the expression to an attitude
which penetrates our whole way of thinking. For deductive proof gives
us absolute cogency an unconditional certainty derived from the la\.,s
of logic. This may seem unfair as l;iarnock is only attempting to be
just, but in giving philosophy of education its due, in setting its
place, she denies that we can have proof and this distorts our whole
view of reasoning. Heasoning need not be deductive and yet can still
be a priori. It need not be inductive either. There can be logical
connections which are ~ priori but non-necessary. This needs to be shown. Perhaps Aristotle had this in mind but expressed it in a
misleading \.,ay? The misleading expression is what we must focus upon.
The assumption conte.ined in this misleading way of putting things in
a (correct) picture of the unassailable nature of deductive reasoning.
This general picture when taken as a model of comparison or paradigm
of reasoning forms the ~asis for the other misapprehensions and confusions. The acceptance of this picture is a habit of thought
with philosophers. They do not, as it
were,
say it to themselves.The second feature is meant to characterise essentialism without
gi ving expression to any pe.rticular form of essentialism. Thus it
does not distinguish behleen \.,hat we can call 'vulgar' essentialism
the idea that there q~certain characteristics that are necessarily true of certain objects, and more sophisticated forms of essentialism which
allow any of a varying but given subset of a range of different
features to be necessarily true of certain objects. Peters once
declared that he did not mind if he put his 'foot on the primrose path
been rife ever since.
(48)
The third, fourth and fifth features are a consequence or are
concomitants of the second. Definition; dictionary definition;
general definition; scientific definition; strong definition; weak
definition; descriptive definition; stipulative definition;
prescriptive definition; programmatic definition; inventive and
non-inventive stipulative definition; descriptive-programrJatic definition,
all these are familiar terms to philosophers of education. Often
they deny that there is any 'dictionary' definition of the term
(49)
'education' • Yet they con,stantly turn to the Oxford English
Dictionaries to extract 'exhaustive categories' of education, (50)
or to distinguish between a 'general' and a 'specific' meaning of the
term. (51) Hore typically they sUTmlarise the 'familiar' objections to beginning their enquiry Hith definitions of central terms and
then to argue that, nevertheles,s, this approach is not unprofitable
d ' 1 " f" t tt t" ' t k t " " t" f" d (52) an some c arl lca ory a en lon 0 ,ey erms lS JUS l le •
A specific objection to using the dictionary 1:1hich is not covered by
the usual philosophical objections is that the dictionary has been
(53)
shmvn in SOrle infaelous cases to be 'Trong. And this is something
which cannot be shown by reference to the dictionary. Concern with
definition results partly from the axiomatic model provi~ed by 1) and the method of 2). This latter feature results in a theory of
meaning. For the method is held to produce 'conceptual truths' \vhich
look very much like definitions of the sort found in dictionaries.
Thus 'to be educated' is said to mean 'to :earn to be a person';
(54)
'Teaching' is said to mean 'the intention to bring about learning in
(56)
another'; and so on. These are said to be 'conceptual truths'
'lhich mark conceptual connections. In a less material mode we are
said to have '(logically) necessary conditions' for the use of the
term 'education'. These provide us 'dith 'criteria' for the use of
terms. Peters' statement of the (three) 'criteria' of education in
Ethics and Education or Hirst and Peters' statement of the (two)
" t " d t " " mh T " f
~d
t " (57) " 11crl erla of e uca lon In 1 e .wOglC 0 .t'., uca lon I'll serve as
examples. The presentation of 'criteria' of this general sort
follows from features 1) -
3).
Often the notion of a 'rule' or 'rules' for the use of the term is introduced to expedite theTerms such as 'criteria' and 'rule' are often explicitly said to
derive from Uittgenstein. 3ven terms "rhich do not are incoruorated
into an interpretation of his Hritings. But the use of these terr.Js in
the \\Titings of i.'Iittgenstein are highly problematic. They are used as
explanatory devices when they are in need of careful explanation
themselves.
The sixth feature, the constant need to reformulate analyses is
(58)
something that Peters comments upon. The seventh feature, the
ongoing argument about method, follows from the sixth point.
Philosophers of education stick to the idea that they have to set out
'conceptual truths' despite the seeming impossibility of producing
any true 'truths', and despite substantial criticism. The critics
fall into two sorts. There are 'internal' critics such as Reddiford
(58)
and Dunlop (60) who ,.,.ish to explore the possibilities of non-analytic,
non-contingent truths connected "'lith the concept of a 'human being'.
i;Je can also include in this group all those critics 1,..rho ,..rant to push
all such truths into the contin,';;ent camp. (61) Then there are
'external' critics such as
Haac~ (6a)an~
Nidditch(63)
who are muchmore uncompromising. Nidditch, for example, rejects the method of
'conceptual analysis' which he calls 'ITeo-Ramism' on three grounds:
Firstly, he objects to an Aristotelian assumption about the uniformity
of hUr:1an beings that the t:J.ethodoloE';y requires. (64) Secondly, he objects to the unargued nature of many of the assertions made by the
practitioners of Neo-Ramasm.
(65)
And finally, he objects to the. (66)
values and preferences of the authors masqueradlng as 'truths'.
\fuatever the validity of these points, \.,.hich have been made by many
besides Nidditch, there is clearly a debate about method.
The final point is a surprising one, as it is often baldly
asserted th8.t one of the cardinal points in philosophical method is to
(67)
show points by means of examples.' This leaves the giving of
examples in philoso"Jh;y of education as no more than a curiosity. It
is an entirely accidental or contingent feature of the subject. As it
stands the remark is ambiguous. No attempt is made to clarify or
expand it. It could refer to the Kantian view "jhich sees examples as
having an important illustrative function. Or it could refer to
some-thing more fundaJ!lent:aL _ That. it does not is apparent when you start
to count examples. Not that there is something in the sheer quantity
is more likely to be indicative of the Kantian view than otherwise.
If an instructive comparison is sought it might be useful to compare
Paul Ziff's discussion of 'The ',\ford "Good'" in Semantic Analysis (68)
with Peters' discussion of education in Ethics And Education. Both
are about forty pages long. Ziff considers in detail over 160 examples of the use of the term 'good'. There is nothing like a
quarter of this number in Peters' section, and to say this is to be
very generous in what we take as an example.
The paradoxical nature of these tendencies is now apparent. The
pull of the mathematical analogy is resisted even in the piece of
documentation produced to support it. Essentialism is decried while
essentialist methods of attempting to see 'behind' our use of terms
are retained. Definitional approaches are substantially criticised
and vigorously attacked while definitions, of a sort, are given.
Universally a~plicable 'criteria' are sought while their multiplicity is recognised. Analyses are constantly altered and reformulated and
the method of 'conceptual analysis' attacked and defended. Examples
are held to be important but are not extensively used. And this
situation is held to be a 'healthy' state of affairs.
All this is not new and 1.·.rill not be news to philoso:phers. It
is only to be expected given a certain view of the task of a
philosopher. This view is that what is characteristic of philosophy
is its 'quest for generality' its seeking after a 'general perspective'.
These remarks come from Scheffler's seminal work The Language of
Education. (70) I know of no evidence that suggests that the
conception of philosoJlr,y of education he gave expression to there has
been overthrown. Indeed, philosophers never tire of reminding us
that 'by its nature, in looking to universal principles, the
(71) philosophy of teaching is abstract as philosophy is always abstract'.
The consequences of this perspective are disastrol.4s when 'vIe move
from the 'healthy' self-reflective monents of the philosopher of
education on his subject to his attempts to solve particular
problems. In a primary school where explanations, in the early years
at least, may often seem to t~~e the form of ostention or exrunple-giving and in this way be 'radically different from explanations in
other institutions', (72) they are seen by one philosppher of
'definitions', 'standards', 'norms', 'criteria' and so on. The use
of examples is not mentioned and there is certainly no indication
that consideration of them is necessary. The use of examples is
implied for instance in the giving of some procedural explanation
of a piece of know-how. But the use of examples is not held to be
worthy of treatment parri passu. The efficacy of a satisfactory
veroal explanation is such that examples seem only an indirect
way of reaching the duller vessels who cannot dispense with them.
The consequences of this view could be the misconduct of education.
For the moment we are concerned with the Kantian malaise at the
most general and academic level. And at this level there is no
evidence that this conception of philosophy has been overthrovm.
The appearance of these tendencies a~d confusions listed above are obvious manifestations of what has been called a 'craving for
generality'. An alternative formulation of this remark would be to say that these philosophers have a 'contemptuous attitude
, (74)
towards the particular case • Expressions such as 'craving'
and 'contempt' are not intended to be insolent or unsavory which
they could seem to be if taken out of context. They are not
superlatives. They indicate precisely that what is at issue is
not a mistaken analysis, a false assumption, or a mistake in
argument, but a way of thi~~ing that is habitual and unreflective. It may involve false assumptions and mistakes at another level but
these are merely SY;l1ptoms of a flawed apprehension. \'lhat follows
may be considered as a plea for particularity.
1.3. Documentation (B)
1:1e have been concerned to document what seems to be a flawed
apprehension of the concept of education amongst philosophers of
education. This flawed apprehension will be the theme of this
essay. Such an activity as discussing a flawed apprehension of a
concept may seem academic, frivolous and remote. This is a
serious accusation and there are two replies to it, one which is
frivolous, and derives from John '.hsdom and runs like this
, ••• it's a free country. Everyman to his taste. Some go to
the dogs and some consider the concept of having the concept of
1,visdom means that \'1e should 'produce evidence that the confusions
th t a concern us are ac t " 1 ve ln non-p h'l 1 oso:9 lca h' I d' lscour .. e. , (
76 )
Upon the production of such evidence depends the meaningfulness
and relevance of my claim that there is something wrong with our
concept of the concept of education. h'e can do no more here than
hint at how we could go about such documentation. It seems to me that the flawed apprehension of the concept of education is
apparent in the following sorts of instances:
(77)
(1) In the existence of demands for the definition of 'education'. Perl1.aps a classic illustration of this is provided by a
letter to The Guardian which announced that as the editor of the 'education' section had not defined what he meant by
'education' people ha
(7
3)
'blank in their minds' as to the meaning of the term.(2) In the existence of the paradoxical situation that it is often claimed that we are all experts on education and yet at the same time we are ignorant of its nature. This sort of argument is regularly produced and commented upon. See for example an article 'Educating the Educators' published in the New Statesman. (79)
(3) In the existence of a popular debate about how to talk about education. 'Popular' in the sense that it is not restricted to academics. Here we could refer to discussions about what goes on in primary schools. (80) Bore specific-ally a group calling for 'Education for Capability' has recently set out to alter our education system at least partly by altering the way we talk about education. They explicitly state that they wish to alter our concept of education not merely to include but to stress 'training'.
~he trend of phllomphy of education has been to stress that since the nineteenth century the concept of education had been progres,si vely separated from that of training. This they would argue is a deba"sement of the concept of education. As evidence of this they point out that we have no word for the culture the Germans describe as
'Tecknik' or for the mode of working the 5'rench describe as 'Metier'. In Emile, of course, the concepts of
education and training are almost synonymous. Of a heter-ogen~ collection of some one hundred and forty signatories to a letter in The Guardian setting out the aims of the group none are philosophers of education. (81)
(4)
In the continuance of a public debate about the nature of education. 'The then Prime lJiinister, Hr. James Callaghan, introduced just such a debate on 18th October 1976. (82)No attempt is made to deal with all the possible permutations of
these instances. ]'.jany other examples could he produced from newspapers,
magazines, television and radio, as 'dell as from everyday life. This
limited gesture towards such documentation demonstrates the extent of
what seemed prima facie a purely academic or philosophical misapprehension.
1.4.
Examples and Counter-ExamplesI have been indicating "'That seems to be evidence of an inadequacy
in the treatment of the concept of education by philosophers of
education. I have called this inadequacy a result of the 'Kantian
vie,,, of examples'. Under this view of examples they have a clear
function. They illustrate a general hypothesis. This belongs to
a whole style of philosophising which is well illustrated by
Scheffler's method. Typically Scheffler makes a few general remarks
d th d t . " 1 t d 1
(84)
All an en procee s 0 conSlaer se ec e examp es. ourattention is focused on the general. 1'·low this is clearly a role
for examples and a useful one. As I have said, the Kantian view is
not all wrong.
This typical way of thinldng has a corollary in the giving of
counter-examples. It is usually held pace the treatment of
scientific hypotheses that one counter-example is sufficient to
refute a thesis.
(85)
Let me illustrate this method with animagine.ti ve counter-exarnple to the epistemological hypothesis
that 'Knowledge is justified true belief'. The counter-example to
this well entrenched thesis was thought up by Colin Radford and
goes as follows: 'Imagine the ca.se of a French-Canadian who believed
he knew nothing of English history. In the course of a game he
makes several correct guesses about dates in iliglish history,
including the date of the death of Queen Elizabeth in
1603.
Atthe time they seem to be merely guesses, but he subsequently
remembers what he had forgotten, namely, that he once had to
learn these dates in school. So he did in fact know the dates,
although he did not believe he did. Giving this sort of
counter-example is clearly a useful role for counter-examples.
,(86)
This brief sketch does not start with an over-simplistic view
of scientific method. What we are presenting is part of a picture
of scientific methodology and one of mythological proportions •. !.,It
could be argued that whatever fails to fit our systems or breaks
scientists to develop new theories. By examining cases of how
scienb.sts actually proceed Feyerabend has convincingly shown
that the gauntlet is not taken up and 'vlhatever fails to fit
into the established category system or is said to be incompatible
with this system is viewed as something quite horrifying or,
more frequently, it is simpl;y declared not to exist.
,
(87)
General theses and counter-examples also fe3.ture largely in
literature on the conceut of education. The trouble is we are
reluctant to give theses up. Just as there have been desperate
attempts to rescue that famous definition of what we mean by
(88)
knowledge. There have been equally desperate attempts to
rescue the infamous definition of 'education' as 'initiation into
a worthHhile way of life'. Obvious counter-examples to this are
someone undergoing an education in a 'Robinson Crusoe' sort of
situation, or someone's saying 'Travelling vlhith him was a real
education.' (89) Desperate
atter~pts
are made to save the general thesis. The concept is battered, re-shaped, cut up, squeezed andforced into the required form. The methodology of this metamorphosis
has been illustrated. But this way of dealing \,ri th counter-examples,
by shuffling them off the stage, is so co;'~mon place that it is rarely questioned. It has indeed been called the 'Classic Hove' of the
(90)
conceptual analyst. Again, there is clearly some validity in
such a move at some tir:Jes in an argument. It CB.n be a way of
focusing someone's attention on the cases you Ivant therr. to consider.
But to persist in making this move is to risk conceptual distortion.
'Examples', 'Instances', and 'Cases' are used as if they were
interchangeable. But there are important differences which we can
indicate in a rough and ready fashion. An example suggests something
imaginative and creative. lde give examples when we are struggling
to understand or to challenge a received underst~~ding as in the Radford example.
An
instance is evidence for a general argument or hypothesis.An
instance must be an instance of something, as in philosophy ""e talk of all instance of some universal or other.'Case' is a more neutral and general tern referring to events,
happenings, instances and examples. Equivocation between 'instances'
Thus to defend the conceptual analyst's belief in the existence of
something 'behind' our use of terms against those who stress the
importance of a consideration of a manifold of cases it is claimed
that
'An
example must be an example of something'. And this is true,but not in the sense intended. liie shall return to this last point.
Examples as they feature in the abundant literature on the
concept of education are cursory, hackneyed and restricted. Cursory
examples are a result of seeking a 'general form' for 'education'
which makes no reference to content. Connections between 'education'
and 'knowledge' and 'understanding' can be made explicit without
committing oneself to any particular content. Thus we get offhand
references to 'experiences being transformed' and to people still
listening and learning even in middle age ~~d when it does them no financial or other extrinsic good. Or to a man who only knew
mathematics or who was trained in military skills and habits.
Hackneyed exrunples abound Spartan Education, Robinson Crusoe,
anamnesis as demonstrated in the Heno, and so on. The lack of
specificity in examples results from the concept itself which demar-ds
the development of knowledge and understanding in breadth. This
restricts examples eve~ more than the restriction on 'content' and presents an illusion of understanding.
A defence of 'trivial' or 'simple' examples seems possible in
philosophy of education as in philosophy of morals or religion. An
argument that education is necessarily a non-trivial thing would have
equal force in the case of these other branches of philosophy where
the suggestion is often made. By 'trivial' philosophers mean to
indicate that the examples must be such as not to enbroil you
immediately in a heated and emotional dispute. To be 'simnle' they
must be transparent in a sense which will be made clear. Philosophers
who consider such examples do not sepa~ate the 'trivial' from the 'simple' but it is useful to distinguish them to meet the obvious
objection that a trivial example is going to be simple, but a simple
example need not be trivial.
The giving of trivial examples in ['loral philosoDhy is held to be
not only necessary but desirable. Thus exwjples such as whether we
ought to return our overdue library books, or pay for our groceries
are held to illuminate the nature of moral judgements because what
we ought to do in these cases is obvious. The suggestion is that
the content of those juagements. This is consonant with the thesis
that moral judgements are unive~isable i.e. that I hold that X is the right thing for me to do in situation S commits me to holding
that it is the right thing for Y to do in a not dissimilar situation.
A consideration of a full and complex example will show that this
thesis is false at least for a class of moral judgements and in doing
so will show the necessity of considering fully fledged examples.
The counter-example comes from Jilel ville • s noveJ2.tle. Billy Budd. (91 ) Captain Vere after hearing the case decides that budd is guilty of
murder and should be put to death. This is not a simple rule-book
decision it involves considerable reflection on Budd's actions, the
si tuation he found himself in on the H M S IndomH::ah£e... A
different Captain or even Vere himself at a different time may
reach a different decision about \IThat is morally right. There
may appear to be no difficulty here. The difficulty becomes
apparent when it is claimed that both judgements can be morally
right. This may seem to establish that moral judgements are
ineluctably particular only at the cost of contradiction and
inconsistency. But this is not the result and we can see this if we
represent the matter schematically. V's decision that X is the
right thing to do in situation S does not commit him to holding that
H in a not dissinilar situation S should also decide that X is
right. For things may weigh quite differently with T,,,'. He is a person
with different inclinations and a different dis:position. This only
comes out when a complex example is considered. It must be pointed
out that the decision is not a problematiC one. The facts of the case
are not in question. It is only in such complex or full-blown
examples that \ITe can see this feature of moral jUdgements. In the
trivial cases we have extracted 'universalisability' as a
character-istic of moral judgements precisely because the cases we are
considering ~ trivial in th2.t they lack exactly that which makes our concerns moral. In the case of 'Starry' Vere this could be said
to be a conflict between a strict naval code and humanitarian
feelings. If we remove this content \ITe are not dealing with a
recognizable moral problem.
Bambrough makes out a strong case for considering 'simple'
examples. He argues that they have two advantages. Firstly they
they are transparent in the sense that they allow us to see the
patterns which connect these cases with more complicated ones.
This is possible because they lie in an absolute unbroken continuum.
He warns that we must not think that simple cases allm..r us to see more
clearly a single element cornmon to all cases, rather that we look for
'patterns of analogy or family resemblance'. (92) Bambrough comments
that his remarks are quite general in their application. Drawing
on arguments familiar to those who have read his influential paper
on 'UniveISa~ and Family Resemblances' Bambrough argues that in the case of religion there must be such patterns of reser:lblance
because vIe do after all call the Gods of Olympus and the God of
Hoses 'Gods'. As 1I1e shall see in Chapter Three Bambrough is repeating
a mistake of his earlier paper. The position he is maintaining is
simply a sophisticated version of the essentialism he disparage~ This explains why so many diselaimers appear in his argument to the effect
that differences are as important or more important than similarities.
But Bambrough also equivocates between 'trivial' and 'si~ple' cases. The examples he gives are far from trivial. Thus he dwells at
length on Homer's account of Athena appearing before Telemachus in
the guise of Nentor. This is not trivial in the way that not
returning your library books is. And indeed Bambrough admits that
his simple examples are not so simple after all. 'Simple' seems to
mean a :oerspicuous or even good example. However we interpret the
appearance of Nentor this interpretation will be coloured by the fact
that we don't believe in these gods. ljJhatever account ,. 1.ole give will
shmw.that we have ~mythologised whatever was intended by the writer. Bambrough then proceeds to sholtl hOlt' certain 'theologians' such as
the Bishop of lj!oolwich have demythologised Christiani ty in a similar
fashion. Jux~aposing such 'si~ple' cases may indeed illuminate but this illumination is not a function of their simplicity.
Examples in philosophy of education have a contrived air.
Consider Johanna Burgess' attempt to construct an 'academic paradox'
out of Peters' analysis of 'education'. She gives the example of a
professor who is doing beneficial 1!lork but receiving payment from a
dubious source something which as a moral man he should reject.
Have we here a case of a man vlho is doing something worthwhile and
educated? And is it an answer to say that this exploits two senses
of 'worthwhile' and that a man can be educated a~d not live up to his ideals?
(93)
Doesn't this just fail to see the force of theparadox in the way that St Paul failed to see the force of the
paradox of the Liar \"rhen he considered the fact that they always lie
just another indication of \"rhat a bad lot Cretians were? 1.vnat are we
to say about this example? It illustrates a further feature of the
kind of examples that 'tIe find in philosophy of education. They are
hopelessly indeterminate and this indeterminacy precludes the
possibili ty of any sound judgeme:'lt. IIore than this, an indeterminate
example lends a spurious credibility to whatever argument is being advanced. (94)
That examples are often cursory, hackneyed-;~ trivial, and indeterminate is a result of the Kantian view. If we primarily
seek 'conceptual truths' of one sort or another, or 'criteria', or
'rules', or 'logically necessary conditions' for the use of terms
then the use of examples Hill remain simply a matter of presenting
'instances' to indicate whatever 'universal' we have in mind. To
drive this point home, consider an analogous case from literature.
Turning our discussion of trivial and poor examples on its head we
may point out that any set of general considerations that establish
that King Lear is sad will do the same for the death of Little Nell,
and, as Oscar Wilde remarked, it \vould talce a heart of stone to read
the latter without laughing.
Exru~ples can be disparaged for non-philosophical reasons. In politics the;yrcan mislead Hhen uroduced as evidence or
counter-evidence for hypotheses requiring quantitative consideration. An
instance would b'2 the suggestion that unemployment benefit is
claimed by people who nearly all have clandestine jobs. Instances
of dramatic abuse may even have effect at the level of social policy
making. The use of exa~ples, analogies and allusions in such cases leads to vagueness and irr:precision. This mayor may not be
consciously contrived. It remains a common misuse of ex~nples
1Ilhich could be fairly listed in some loose pb:i:LloSJ]hical categorisation
such as Bentham's political fallacies among other'arguments' used by
Another danger in giving examples in the eyes of educationalist
Herbert Kohl is that people use them as mod~ to imitate. An example becon:es an exemplar. A familiar danger mentioned by Wi ttgenstein
is that one feeds one's mind with 'a one-sided diet' of examples of a
single kind (P.I. s
593). (97)
This remark is ,'lOrth re"oeating as one h " ,- h h t d -'-h t . t ' d' t -d t' (98)p LwSOP er as sugges e "a 1 lS a 18 ary recommen a 10n.
Again examples can be seen as something akin to the phenomenologist's
'pure situation' which you have to consider carefully if you wish to
th 1 . f . th,'
~
(99)
Th d f 1see e rea meClnlng 0", sO,l1e ---nES. ese angers 0 exarnp es
result from the Kantian view. In the first case we simply generalise
from a single instance in a classically fallecious ',ay. In the ,second
case He give credence to one particular view by selecting only examples
which exemplify that perspective. In the last case 'lIe have simply
another variant of essentialism.
The Kantian view leaves us vJith a distorted practice. We give
examples in ways ,,,hich exhi bi t our prior concern vIi th more general
features of our concepts. CritiCising this view does not mean that
we should simply consider more counter-examples or borderline cases.
To do this would be to invoke in philosophy of education something
analogous to Feyerabend's 'counter inductive' procedure. This
involves the construction of hypotheses inconsistent with both
accepted theories and established facts. This is necessary in
science because a counter example to an orthodox theory is only seen
as a counter-example i ... e. a refutation if it is enmeshed \"ithin a
theoretical system. This means that 1-1e remain at the level of theory (100) construction no matter hot". \'Ie increase and maximise discrepa."'1cies.
Therefore the 'anti-methodolo~J' of 'anything goes' ,vill not serve to \"eaken the misap:9rehension brought about by the Kantian view.
It has been claimed that this view involves a contempt for the
particular case. At its most extreme it errs in not even considering
the giving of exaT!l1Jles a preli:nL'lary way of answering questions about (101)
-a concept. Or, if this is granted, it is granted grudgingly.
For example, Elizabeth Hindess in her article 'Teaching the l'~eaning of Hords' discussed the questio:l of what it is to know the meaning of a
\,'ord. She allO\".s that there are different senses in which '.-Je can give
demanding' senses in which we can give an account of a lrJord' s meaning:
'It could be no more than giving examples of how the word is used on (102)
any occasion.' This is a clear instance of the Kantian prejudice.
Presumably the more demanding way of giving an account of a word's
meaning would be to produce 'criteria' or 'definitions'. In this essay
I want to correct this sort of misapprehension about the role of
examples. I ,;rant to stress their fundamental importance in giving
an account of a concept. The view expressed by Hindess is that the
giving of examples is an indirect or barely adequate way of explaining
the meaning of a term. Our knowledge of what it means is something
unformulated, perhaps a definition. The hard work comes in formulating
this 'defini tion '. Against this I want to argue that the term
'concept' is vague and that to mal-ce clearer '''hat we mean when we talk
of any particular concept including the cor..cept 'concept', 'de must
proceed by examples. This is how we make things more determinate.
To seek clarification of any other sort '''ill only get us into trouble. (103)
In the Philosophical Investigations Hittgenstein presents a critique
of the Kantian view. So much has been written and said about this that
VJisdom and others have had to warn us that a distaS-te for simplification,
for definitions and the like can be too simpliste. \~Te shall see why later. For the moment, let us take up once again the 'question' that
\'Ji ttgenstein asks his interlocutor in just the sort of situation I have
been discussing:
'Isn't my knowledge, my concept of a game, completely expressed in the explanations that I could give? That is, in my deseribing examples of various kinds of game; shewing how all sorts of other games can be constructed on the analogy of these; saying that I should scarcely include this or this among games; and so on.' (P.I. s.75)
It should be noted that it is not asserted that the giving of examples
is sufficient. :Nor is this denied. A consideration of \rIi ttgenstein' s
style lrJould reveal that the suggestion is entertained or exhibited
rather than asserted. r:::'hus it is not open to a facile form of 'refutation'
that has become commonplace. But for the moment we ,,,ill take it at face
value. There are dangers of misinterpretation that result from doing
this. To see the passage as more assertive than its form suggests may
avoid the Kantian malaise at the risk of appearing to begin to justify
the dangers are minimal and the prophylactic possibilities considerable.
This qualification aside we can put positively the argument that
examples are absolutely fundamental to thought. 7his thesis is not an
original one but i t i~ entirely misunderstood at least by philosophers of education. It derives from the l.jork of John Wisdom who in the
closing paragraphs of his article 'A Feature of \iittgenstein's
Technique' puts the corrective to Kant's view succinctly:
'Kant said that examples are the go-cart of the human understanding. But this is not enough. Examples are the final food of thought. Principles and laws may serve us well. They can help us to bring to bear on what is now in question what is not now in question. They help us to connect one thing with another and another. But at the bar of reason, ah/ays the final appeal is to cases.'
Vlisdom's remark, though colloquial and simple in its presentation, is
laconic and difficult. He had developed his ideas about the role of
'examples' or 'Instances' or 'cases' in his
1957
lectures on 'Proofand Explanation' delivered at the University of Virginia. In these
lectures, cOf!lmonly known as The Virginia Lectures, he defends in
detail the idea that examples are fundamental to thought. In doing
so he utilises the notion of what he calls 'Case by Case Procedure'
or 'Case by Case Argument'. I-.'hat I have to say about the importance
of examples derives entirely from a careful reading of these lectures
supplemented by consideration of Hisdom's other writings and of the
relevant work of \ii ttgenstein. As the lectures are not widely
available, I will sumr:larise the most relevant passages and arguments
before going on to apply them in an effort to rectify the inadequate
apprehension of the concept of education which exists in the recent
literature.
(105)
1.5.
Case By Case ArgumentThe aspect of the Virginia Lectures that I want to concentrate
upon is viisdom' s claic1 that 'j,ll reflection comes in the end to a
case by Case procedure.' (VL. IX. 1.) Time after time he suggests
that failure to recognise just this point is the root cause of many
freely available su'nmary of the Lectures, remarks that this 'does (106)
not need to be laboured.' This dismissive attitude is
thoroughly wrong-headed. In this account of what I believe to be
h'isdor:1's central theme I want to stress how radical and original this
thesis is.
Passmore describes Hisdom's philosophical method as consisting
in his 'first making a distinction •••• as if it were a sharp one ( 107)
and then blurring its edges.' This puts him at odds with Kant
and contemporary philosophers like Quine. For example, Kant sought
for a 'criterion by which to distinguish ,.nth certainty between
d "" al kn 1 d , ( 108) I ""1 . pure an emp~r~c ow e gee n a s~m~ ar manner ~~ne
cri ticises empiricists for their metaph::Tsical belief in the
analytic/synthetic distinction because 'a boundary behJeen analytic
and synthetic statements simply has not been drawn.' (109)
~uite
clearly r:',ruine is working with Kantian assumptions. The existenceof borderline cases seems to show, as far as he is concerned, that
we cannot distinguish with certainty between the two sorts of
statement and therefore the distinction is a 'dogma'. i"lisdom is
not troubled by such demands for 'certainty'. Is a satyr a man or
a goat? He may hesitate over the answer 'but that doesn't mean
that there is no difference between a goat ~~d a man. So if it is the fact that we hesitate over borderline cases that is being
referred to, I shouldn't put it in the form that a boundary has
not been drawn.'
Wisdom's method may seem like Quine's but i:)line takes the
existence of borderline cases to show that the distinction is a
metaphysical dogma. Wisdom, on the other hand, questions the
underlying Kantian assumptions at l'lOrk when this sort of move is
made. He shows that such demands for 'certainty' must be rejected
because they embody a false picture. Nor is Hisdom introducing
the concept of 'broad borderlines' as a modification of the Kantian
viel-.r. 'rhis interpretation, though tenable, and seemingly supported
b Y passages ~n " h V" t e ~rg~n~a " . L t ec ures, ( 110 ) . ~s one th t t l a mus u -,_~ma
~
. t 1 e ybe rejected. l;lisdom' s point goes much deeper and challenges a whole
